Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 13890
    Prev
    Next
    INDEX
    Explanations

    legal terminology related to motions, filings, and evidence in a judicial context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    legal or technical procedures

    np_acts-logits-general · gemini-2.0-flash

    notice evidence modify transmit estimating procedures assess

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    AddTagHelper
    -0.55
    lccn
    -0.53
    TagMode
    -0.53
     endregion
    -0.53
    flashdata
    -0.50
     <<<<<<<<<<<<<<
    -0.48
    DebuggerNonUser
    -0.48
    EndProject
    -0.47
    SupportActionBar
    -0.47
     vlieg
    -0.47
    POSITIVE LOGITS
     diagon
    0.40
    openzeppelin
    0.36
    WithIOException
    0.35
     BoxDecoration
    0.35
     consultation
    0.34
     communi
    0.31
     CONSULT
    0.31
    ükemmel
    0.30
     abus
    0.30
    Consultation
    0.30
    Activations Density 1.080%

    No Known Activations